[lowrisc-dev] Open GPU for the first CPU

Reinoud Zandijk reinoud at NetBSD.org
Fri Feb 13 17:19:58 GMT 2015

On Sat, Feb 14, 2015 at 02:13:53AM +1100, Jookia wrote:
> On 02/14/2015 01:56 AM, Reinoud Zandijk wrote:
> >Not to stir a flamewar, but his GPU hardware code is released under LGPL which
> >might be feasable.
> Not wanting to stir a flame war either, but if there's only one option for
> an open GPU in the end, perhaps it'd be better to have a branch that's
> LGPL/GPU with the GPU until a permissive GPU is viable?
> I'd hate to see an open GPU not used due to it's license.

True, but one has to draw a line somewhere; LGPL is usable but GPL isn't. This
is since stuff labeled with LGPL can be linked/used in say a BSD or whatever
(compatible) license where as GPL can't since it can only be used in an
all-GPL environment.

> That said from a practical standpoint it'd be nice to outline what
> performance benchmarks a GPU should achieve to be deemed usable. Desktop
> compositing at minimum, though playable Quake would be nice.
> I'm probably vastly underestimating how unfinished and low performance the
> GPU would be, but I'm just amazed it exists at all.

A GPU is not very much unlike a `normal' CPU but only focusses on different
things. Interrupts f.e. are not really needed, context switching neither. What
is needed is a good amount of hardware threads and a good amount of CPUs to
try to maximise memory bandwidth and minimize delays.

As for lowRISC, the memory bandwidth to the DRAM is, as it looks seems now,
quite good and even a `dumb' framebuffer could do compositing. I'd rather have
a good working `dumb' framebuffer than one fraud with issues. Would be a nice
development for lowRISCv2 :)


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: not available
Url : http://listmaster.pepperfish.net/pipermail/lowrisc-dev-lists.lowrisc.org/attachments/20150213/a42bd3da/attachment.sig

More information about the lowrisc-dev mailing list